GigaIO partners with Mushroom Networks to solve Near-edge to Far-edge connectivity challenges

Patented bonding technology removes connectivity bottlenecks by providing better communication links, significantly greater bandwidth, and seamless failover between sites.

GigaIO, a pioneer in delivering datacenter-class computing at the edge, has partnered with Mushroom Networks, inventor of Broadband Bonding, a patented, SD-WAN solution that aggregates multiple wireless and wired connections — including Wi-Fi, cellular, private 5G, fiber, and satellite — into a single high-bandwidth, self-healing link with seamless failover. This partnership produces a solution that provides very robust, high-bandwidth connectivity — wireless, wired, or a combination thereof — between GigaIO’s edge supercomputers and other compute infrastructure.

AI at the edge is only as good as the data you can get to it. Edge deployments generate enormous volumes of sensor and video data, but the links connecting the far edge to compute resources are often the weakest part of the chain — wireless, bandwidth-constrained, and prone to failure. Traditionally, the low bandwidth provided by wireless connections can be problematic, particularly for video data.

By bonding multiple wireless and wired connections into a single high-bandwidth, self-healing pipe with seamless failover, Mushroom Networks’ technology ensures that data flows to GigaIO’s edge compute platforms reliably, whether in a factory, in the field, or in an emergency deployment. These capabilities are particularly important to enable far-edge to near-edge communications, where in many environments this link needs to be wireless because making wired connections would be impractical.

Such connectivity can enable rapid set-up of communications in emergency situations, providing for multiple wireless connections as well as a mix of wireless and wired links that seamlessly bond multiple channels into one fast, resilient, secure connection to increase bandwidth. The system automatically monitors individual connections and manages failover situations where one or more of the connections fail, and is very effective at controlling communication between infrastructure located in different sites, utilizing dedicated tunnels between boxes so that communication is protected. Pairing this connectivity with GigaIO’s edge compute platforms means teams can stand up datacenter-class AI in the field in minutes, with the robust communications to match.

“Increasing communication performance and robustness is always important, so we are excited to work with Mushroom Networks to dramatically help solve near-edge to far-edge connectivity challenges,” said GigaIO CEO Alan Benjamin. “As sensor data grows, moving that data to where it can be processed in the time frame required is growing more challenging. Together with Mushroom, GigaIO can both increase compute power locally and move data efficiently — a powerful combination for Manufacturing 4.0 and 5.0 environments.”

“Sensor and video data at the far edge is growing faster than the networks that carry it,” agreed Mushroom Networks CEO Jay Akin. “Our patented bonding technology turns multiple unpredictable connections into a single predictable, high-bandwidth link with seamless failover — exactly what latency-sensitive AI workloads demand. Combining Mushroom’s intelligent connectivity with GigaIO’s edge compute power gives Manufacturing 4.0 and 5.0 environments a complete edge solution: process more data locally, and move it reliably when you need to.”

GigaIO’s Gryf is the first-ever carry-on suitcase-sized AI supercomputer, built to bring datacenter-class computing wherever it is needed. Manticore is GigaIO’s non-mobile, half-rack-sized unit. Both are available now with Mushroom Networks connectivity options.

Source: BUSINESS WIRE

Related News.

Subscribe to our newsletter!

if you dont want to swim alone in the ocean of news, sign up for the newsletter, and you will receive daily all the important news of world shipping!

* indicates required
Consent *
By submitting this form you agree to receive Email Marketing

Design & Development by P.KAN.DESIGNER

Design & Development by P.KAN.DESIGNER

Privacy Preference Center